The mouse mammary tumor virus (MMTV) induces mammary carcinomas and, at lower frequency, T-cell lymphomas. These lymphomas contain clonal acquired copies of MMTV proviral DNA with LTR deletions. The deletions invariably remove a negative regulatory element (NRE) that is responsible for suppressing MMTV transcription in tissues other than the mammary gland as well as the 3' one-third of the sag gene that encodes the viral superantigen. The sag gene recently has been shown to be required for the proliferation of T-cell intermediates involved in the transmission of milk-borne MMTV from virus-infected cells in the gut to the target cells in the mammary gland. I have developed a model for MMTV-induced T-cell tumors in which loss of NRE sequences gives higher MMTV expression and resultant proviral integrations in T-cells. Moreover, concomitant loss of the C-terminal region of the Sag protein results in the failure to transmit MMTV from T-cells to mammary gland cells, and an unregulated proliferation signal to the T-cell. Various aspects of this model will be tested in this grant proposal. In the first specific aim, we will explore the molecular basis for NRE function in different mouse tissues. The effect of mutations at several NRE binding sites will be characterized further, and NRE mutations will be analyzed in vitro in transient transfections and in vivo in transgenic mice for their effect on tissue- specific expression. cDNA cloning of NRE binding proteins (NBPs) will serve as means to develop NBP antisera that will be used for the identification of NBPs and associated proteins in different tissues. In the second specific aim, we will analyze the role of the truncated Sag protein in MMTV-induced T-cell tumors. Mutations that inactivate Sag or mimic naturally occurring T-cell specific sag deletions will be tested for their ability to induce T-cell proliferation in mixed lymphocyte culture in vitro and to affect thymotropic MMTV oncogenicity in vivo. The ability of thymotropic MMTV to infect the mammary gland will be assayed. MMTV variants that have mutations in sag or the NRE will be tested for tumor- inducing capability. In the third specific aim, we will identify common target sites for MMTV integration in T-cell tumors. We will screen a variety of MMTV-induced T-cell lymphomas for the presence of MMTV insertions at integration sites cloned from a T-cell tumor containing four acquired MMTV proviruses. This system represents a unique opportunity to dissect the molecular nature of disease specificity in two highly related strains of virus. Moreover, our ability to use in vitro systems in combination with transgenic mice provides us with the tools to dissect MMTV transcriptional elements and coding functions at the organismal level. This work has important implications for understanding virally induced cancers and leukemias.